JP4487513B2 - Object recognition device - Google Patents

Description

  The present invention relates to an object recognition apparatus that recognizes the state of a workpiece by performing image processing on an imaging result of an object that moves relative to the imaging unit, such as a workpiece on a conveyor, and particularly to an imaging unit at an imaging timing. The present invention relates to an object recognition apparatus that can obtain a positional relationship with a workpiece with high accuracy and obtain a highly accurate image processing result.

2. Description of the Related Art Conventionally, techniques for recognizing a state of a workpiece by processing an image obtained by imaging the workpiece by the imaging unit while one or both of the workpiece and the imaging unit are moving are used in various industrial fields. ing. Such a technique is used in work inspection work, sorting work, assembly work, and the like in a factory production line (see, for example, Patent Documents 1 and 2).
FIG. 7 shows a typical example of the prior art.
When the workpiece 600 flowing on the conveyor 500 is detected by the trigger sensor 400, a workpiece arrival signal S40 is transmitted to the image processing apparatus 800, and the image processing apparatus 800 uses the workpiece arrival signal S40 as a trigger to cause the imaging means 200 to perform. The result of imaging the workpiece 600 is processed by the image processing apparatus 800 to recognize the state of the workpiece.

FIG. 8 is a block diagram for explaining a conventional technical example.
The image processing apparatus 800 includes a trigger sensor 400 and an imaging unit 200. The image processing apparatus 800 includes an image capturing unit 820, an image memory 130, and a microcomputer (hereinafter referred to as a microcomputer) 140. .
The operation of the object recognition apparatus configured as described above will be described.
(1) The trigger sensor 400 detects that the workpiece 600 has reached a position where the imaging unit 200 can capture images, and outputs a workpiece arrival signal S40 to the image capturing unit 820.
(2) The image capturing unit 820 instructs the imaging unit 200 to start imaging the workpiece 600 by outputting the imaging start signal S41 to the imaging unit 200 using the workpiece arrival signal S40 as a trigger.
(3) The imaging unit 200 outputs the imaging result of the workpiece 600 to the image capturing unit 820 as an analog video signal S10.
(4) The image capturing means 820 internally converts the video signal S10 into a digital image signal S24 and temporarily stores it in the image memory 130.
(5) When the image capture means 820 completes the temporary storage of the required amount of images, it outputs an image capture completion signal S27 to the microcomputer 140.
(6) Upon receiving the image capture completion signal S27, the microcomputer 140 reads the image signal S25 from the image memory 130 and performs image processing.

JP 2000-241363 A JP 2002-259943 A

However, in the conventional object recognition device, the workpiece arrival signal S40 of the trigger sensor 400 is a trigger, and the imaging unit 200 instructs the start of imaging of the workpiece 600 via the processing in the image processing device 800. A time lag that cannot be ignored occurs between the arrival timing and the imaging timing, and it is difficult to recognize the state of the workpiece with high accuracy.
Further, when a general CCD camera is used as the imaging means 200, the imaging start signal S41 cannot be received during the vertical synchronization period due to the characteristics of the CCD camera, and therefore it is difficult to predict the time lag. there were.
According to the present invention, an object recognition apparatus that processes an imaging result of a workpiece that moves relative to an imaging unit and recognizes the state of the workpiece accurately detects a positional relationship between the imaging unit and the workpiece at an imaging timing. An object of the present invention is to provide an object recognition apparatus capable of obtaining a highly accurate image processing result.

An object recognition apparatus according to a first configuration of the present invention includes an imaging unit that images a workpiece and outputs a video signal, a trigger sensor that is used to generate an imaging timing of the workpiece, and an image processing that processes an imaging result of the workpiece In the object recognition apparatus that recognizes the state of the workpiece by performing image processing on the result of imaging the workpiece by the imaging unit while either one or both of the workpiece and the imaging unit move, An encoder that detects a positional relationship between the workpiece and the imaging unit at an imaging timing by the imaging unit is provided, and the image processing apparatus is configured to perform the operation based on the positional relationship between the workpiece and the imaging unit detected by the encoder. It is characterized by recognizing the state of the workpiece.
In this first configuration, the encoder detects the positional relationship between the workpiece and the imaging means, and the image processing apparatus recognizes the workpiece state based on the positional relationship. Thereby, the positional relationship between the imaging means and the workpiece at the imaging timing can be detected with high accuracy.

In the object recognition apparatus according to a second configuration of the present invention, in the first configuration, the encoder outputs an encoder signal generated with relative movement between the workpiece and the imaging unit, and the image processing It supplies to an apparatus, It is characterized by the above-mentioned.
In this second configuration, since the output signal of the encoder is supplied to the image processing apparatus, the image processing apparatus uses the encoder signal to accurately detect the relative positional relationship between the workpiece and the imaging means. can do.
In the object recognition apparatus according to a third configuration of the present invention, in the first configuration, the trigger sensor generates an imaging position initialization signal for initializing a relative position between the workpiece and the imaging means, The imaging position initialization signal is supplied to the image processing apparatus.
In this third configuration, since the output signal of the trigger sensor is supplied to the image processing device, the image processing device initializes the relative positional relationship between the workpiece and the imaging means using the trigger signal. Can do.
In the object recognition apparatus according to a fourth configuration of the present invention, in the first configuration, the imaging unit is synchronized with a vertical synchronization signal generated inside the imaging unit, regardless of a relative position between the workpiece and the imaging unit. At this timing, the result of imaging the workpiece is output as the video signal, and the video signal is supplied to the image processing apparatus.
In the fourth configuration, the imaging unit only has to supply the imaged result to the image processing apparatus as a video signal regardless of the relative position between the workpiece and the imaging unit. The configuration is simple.

In the object recognition device according to a fifth configuration of the present invention, in the first configuration, the image processing device includes the video signal, the imaging position initialization signal, the encoder signal, and the imaging position minimum signal. The image capturing start signal, the image capturing position detecting means for generating the image capturing position signal, and the image capturing start signal are received and the video signal is converted into a digital format and output as an image signal. Then, upon completion of a series of image capture operations, an image capture unit that outputs an image capture completion signal, an image memory that temporarily stores an image signal output from the image capture unit, and the image capture completion signal are received. And a microcomputer that processes the image stored in the image memory using the imaging position signal. To.
In the fifth configuration, the video signal, the imaging position initialization signal, the encoder signal, the imaging position minimum signal, and the imaging position maximum signal are input, and the generated image capture start signal and the imaging position signal are used as input. Since the image temporarily stored in the image memory is processed, the time lag between the workpiece arrival timing and the imaging timing can be reduced.

In the object recognition device according to a sixth configuration of the present invention, in the fifth configuration, the imaging position detection means receives the video signal, outputs a vertical synchronization signal by a known vertical synchronization separation circuit, and An encoder signal and the imaging position initialization signal are input, and a relative movement amount between the workpiece and the imaging means from a preset reference position is generated as a position signal, and the position signal is larger than the imaging position minimum signal. And if it is smaller than the imaging position maximum signal, if the condition is satisfied, the passage of the vertical synchronization signal is permitted, and if the condition is not satisfied, the passage of the vertical synchronization signal is prohibited. An image capture start signal is generated and output from the obtained signal, and the result of temporarily storing the position signal is output as an imaging position signal using the image capture start signal as a trigger. The features.
In the sixth configuration, the position signal is input, the image capturing position minimum signal and the image capturing position maximum signal are used as the determination conditions, and it is determined whether or not to use the vertical synchronization signal as the image capture start signal. Since the position signal obtained by temporary storage is used as the imaging position signal with the image capture start signal as a trigger, an image with a small time lag between the workpiece arrival timing and the imaging timing can be obtained.
In the object recognition device according to a seventh configuration of the present invention, in the fifth configuration, the imaging position minimum signal and the imaging position maximum signal are manually set values such as the microcomputer or a dip switch. It is arbitrarily set by the changing means.
In the seventh configuration, the imaging position minimum signal and the imaging position maximum signal can be arbitrarily set by a microcomputer or a dip switch, so that a flexible system can be constructed.

An object recognition device according to an eighth configuration of the present invention is the object recognition device according to the first configuration, wherein the workpiece is a thin substrate housed in a thin substrate housing cassette, A lid opening / closing device that opens and closes the lid, and a fixing portion of the lid opening / closing device includes a fixing base on which the thin substrate storage cassette can be mounted, a dog mounted on the fixing base, and a linear mounted on the fixing base A moving part of the lid opening / closing device holding a lid of the thin substrate storage cassette, and a lid holding part on which the imaging means and the image processing device are mounted, and a lid holding part. A mounting plate that moves up and down in conjunction with the trigger plate, a trigger sensor mounted on the mounting plate, and a linear encoder mounted on the mounting plate, A thin substrate accommodated inside the mold substrate storage holder, the result of imaging by the imaging means, the image processing by the image processing device, and recognizes the state of the thin substrate.
In this eighth configuration, the imaging means of the object recognition device of the present invention is mounted above the lid holding portion that can be moved up and down, and further, on the mounting plate that moves up and down in conjunction with the lid holding portion, Since the trigger sensor and linear encoder are mounted, and the dog and linear scale are mounted on the fixed base, the image of the state of the thin substrate housed in the thin substrate housing cassette is accurately captured by the imaging means. It can be recognized on the basis.

An object recognition apparatus according to a ninth configuration of the present invention is the object recognition device according to the eighth configuration, wherein the imaging unit is configured to receive a thin substrate accommodated in the thin substrate accommodation cassette along with a lifting operation of the lid holding portion. The accommodation state is imaged, and the result is supplied as a video signal to the image processing apparatus.
In the ninth configuration, as the lid holding unit moves up and down, the image picked up by the image pickup means is supplied to the image processing device, so that the state of the thin substrate in the thin substrate storage cassette can be recognized by the image processing device. .
In the object recognition device according to a tenth configuration of the present invention, in the eighth configuration, the lid holding unit holds the lid of the thin substrate storage cassette removed by the opening / closing mechanism by a mechanism such as air adsorption, It is characterized by moving up and down by a motor drive mechanism or an air drive mechanism.
In the tenth configuration, the lid holding unit moves up and down with the imaging means mounted upward and with the lid opened by the opening / closing mechanism, so that the thin substrate in the thin substrate storage cassette is moved up and down. The state can be imaged.
In the object recognition apparatus according to an eleventh configuration of the present invention, in the eighth configuration, the trigger sensor is configured to move the dog and the trigger at a reference position set in advance when the lid holding unit is moved up and down. The detection slits of the sensors intersect each other, and the trigger sensor supplies an imaging position initialization signal to the image processing device at a timing when it passes the dog.
In the eleventh configuration, when the trigger sensor mounted on the mounting plate that moves up and down in synchronization with the lid holding portion is mounted, the dog and the detection slit of the trigger sensor intersect at a preset reference position. Therefore, the image processing apparatus can initialize the relative position of the lid holding portion with respect to the thin substrate inside the thin substrate accommodation cassette based on the output signal of the trigger sensor.
In the object recognition device according to a twelfth configuration of the present invention, in the eighth configuration, the linear encoder crosses each slit of the linear scale when the lid holding unit is moved up and down. An encoder signal whose state changes whenever the linear encoder crosses each slit of the linear scale is supplied to the image processing apparatus.
In the twelfth configuration, since the linear sensor mounted above the lid holding portion crosses each slit of the linear scale when the lid holding portion is moved up and down, the image processing apparatus Based on the output signal of the encoder, it is possible to detect the relative position of the lid holding portion with respect to the thin substrate inside the thin substrate housing cassette.

According to a thirteenth configuration of the present invention, in the eighth configuration, the imaging position detection means includes the sixth configuration, and the microcomputer uses the imaging position detection means as an imaging position minimum signal. , A value corresponding to the lower height of each thin substrate is set, and a value corresponding to the upper height of each thin substrate is set as the imaging position maximum signal of the imaging position detecting means, and the imaging position The difference between the minimum signal and the maximum imaging position signal is set to an appropriate value in consideration of the frame rate of the imaging means, the field of view of the imaging means, and the relative movement speed of the imaging means and the thin substrate storage cassette. It is characterized by doing.
In the thirteenth configuration, the microcomputer sets the minimum image position signal and the maximum imaging position signal, and the set value is the difference between the minimum imaging position signal and the maximum imaging position signal. The frame rate, field of view, imaging means, and relative speed of the thin substrate storage cassette are set in consideration of the frame rate, the image processing and recognition processing of the thin substrate can be performed by changing the microcomputer program. Can be flexible.

According to the present invention, in the object recognition apparatus that recognizes the state of the workpiece by performing image processing on the imaging result of the workpiece that moves relative to the imaging means, the positional relationship between the workpiece and the imaging means at the imaging timing by the imaging means is determined. An encoder to detect is provided, and the image processing apparatus recognizes the state of the workpiece based on the positional relationship between the workpiece detected by the encoder and the imaging means, so that image processing is performed using the workpiece arrival signal of the trigger sensor as a trigger. Compared to the prior art that was configured to instruct the imaging start of the workpiece imaging via the internal processing of the device, the positional relationship between the imaging means and the workpiece at the imaging timing can be detected with high accuracy, and the image processing result with high accuracy Can be obtained.
In the prior art, the image pickup means has to have a function of receiving a trigger signal. However, in the present invention, a standard CCD camera or the like having no function of receiving a trigger signal is used as the image pickup means. There is also an effect that it is possible.

(First embodiment)
First, the configuration of the first embodiment of the present invention will be described with reference to FIG. 1 and FIG.
The object recognition apparatus according to the present embodiment includes an imaging unit 200 that captures an image of a workpiece 600 and outputs a video signal S10, a trigger sensor 400, an encoder 300 that detects the movement of the conveyor 500, and the image processing apparatus 100.
The imaging unit 200 captures the result of imaging the workpiece 600 as a video signal S10 at a timing synchronized with a vertical synchronization signal generated inside the imaging unit 200 regardless of the relative position between the workpiece 600 on the conveyor 500 and the imaging unit 200. The image data is supplied to the image processing apparatus 100.
The trigger sensor 400 can use either a proximity switch such as a photo sensor or a magnetic sensor, or a contact type switch such as a limit switch, and detects that the workpiece 600 has reached a preset reference position. The imaging position initialization signal S12 for initializing the relative position between the workpiece 600 and the imaging means 200 is generated and supplied to the image processing apparatus 100.

The encoder 300 is connected to a motor (not shown) that drives the conveyor 500, generates an encoder signal S <b> 11 as the work 600 moves on the conveyor 500, and supplies the encoder signal S <b> 11 to the image processing apparatus 100.
The image processing apparatus 100 receives the video signal S10, the image capturing position initialization signal S12, the encoder signal S11, the image capturing position minimum signal S21, and the image capturing position maximum signal S22, and receives the image capturing start signal S23 and the image capturing position signal. When the imaging position detection unit 110 that generates S26 and the image capture start signal S23 are received, the video signal S10 is converted into a digital format and output as an image signal S24. When a series of image capture operations is completed, the image capture completion signal When the image capturing unit 120 that outputs S27, the image memory 130 that temporarily stores the image signal S24 output from the image capturing unit 120, and the image capturing completion signal S27 are received, the image memory is used by using the imaging position signal S26. And a microcomputer 140 that processes the image stored in 130. There.
The microcomputer 140 sets the imaging position minimum signal S21 and the imaging position maximum signal S22 to appropriate values in consideration of the frame rate of the imaging unit 200, the field of view of the imaging unit 200, and the operating speed of the conveyor 500. .

Next, the internal configuration of the imaging position detection unit 110 will be described with reference to FIG.
The vertical synchronization separation circuit 111 receives a video signal S10 input from the outside and generates a vertical synchronization signal S31 that is an imaging timing. The counter circuit 112 counts the number of pulses of the encoder signal S11 input from the outside, and generates the position signal S30 by clearing the count value at the input timing of the imaging position initialization signal S12. The comparison circuit 113 outputs a positive logic signal S32 that is in an active state when the position signal S30 is larger than the imaging position minimum signal S21 input from the outside. The comparison circuit 114 outputs a positive logic signal S33 that is in an active state when the position signal S30 is larger than the imaging position maximum signal S22 input from the outside.
The logic circuit 117 functions as a 3-input AND circuit that receives the signal S32, the signal obtained by logically inverting the signal S33 by the logic inverting circuit 116, and the signal obtained by logically inverting the vertical synchronization signal S31 by the logic inverting circuit 115. Then, the image capturing start signal S23 is generated and output to the outside.
The latch circuit 118 temporarily stores the value of the position signal S30 using the image capture start signal S23 as a trigger and outputs the value as an imaging position signal S26 to the outside.

Further, a timing diagram of the embodiment of the present invention will be described with reference to FIG.
The video signal S10 output from the imaging unit 200 is output in synchronization with the vertical synchronization signal S31, the imaging position signal S26 is updated at the rising timing of the image capturing start signal S23, and then the image captured by the image capturing unit 120 is updated. Writing of the signal S24 to the image memory 130 is started, and an image signal for one field is stored in the image memory 130, and then an image capture completion signal S27 is output.
Here, it is clear that the image capturing start signal S23 and the imaging position signal S26 are generated almost simultaneously with the vertical synchronization signal S31 that is the timing at which the workpiece is actually imaged.

(Second Embodiment)
Next, a second embodiment of the present invention will be described with reference to FIG.
FIG. 5 is a side view of the lid opening / closing device of the thin substrate accommodation cassette according to the second embodiment of the present invention. In this embodiment, the workpiece to be recognized is a thin substrate housed in a thin substrate housing cassette. This thin substrate storage cassette includes a lid opening / closing device that opens and closes the lid 703.
The fixing part of the lid opening / closing device includes a fixing base 711 on which a thin substrate housing cassette can be mounted, a dog 712 attached to the fixing base 711, and a linear scale 713 attached to the fixing base 711.
The moving part of the lid opening / closing device includes a lid holding part 721, the imaging means 200 and the image processing apparatus 100 mounted above the lid holding part 721, an attachment plate 722 that moves up and down in conjunction with the lid holding part 721, The transmission type trigger sensor 410 is mounted on the mounting plate 722 and the linear encoder 310 is mounted on the mounting plate 722.
The thin substrate accommodation cassette includes a frame 701 that forms the outer shape of the thin substrate accommodation cassette, a lid 703 that is opened and closed by a lid opening / closing device, and a plurality of slots 702 that are equipped for the purpose of orderly accommodating the thin substrate 704. Thus, the same number of thin substrates 704 as the number of slots 702 are accommodated inside.

The imaging means 200 is mounted above the lid holding portion 721, and images the accommodation state of the thin substrate 704 accommodated inside the thin substrate accommodation cassette as the lid holding portion 721 moves up and down, and the result is video. The signal S10 is supplied to the image processing apparatus 100.
The lid holding unit 721 holds the lid 703 of the thin substrate housing cassette removed by an opening / closing mechanism (not shown) by a mechanism such as air suction (not shown) and can be moved up and down by a motor driving mechanism or an air driving mechanism (not shown).

The transmission type trigger sensor 410 is configured such that the dog 712 and the detection slit of the transmission type trigger sensor 410 intersect at a preset reference position when the lid holding unit 721 moves up and down. 410 supplies the imaging position initialization signal S12 to the image processing apparatus 100 at the timing of passing the dog 712.
The linear encoder 310 intersects with each slit of the linear scale 713 when the lid holding portion 721 moves up and down, and each time the linear encoder 310 intersects with each slit of the linear scale 713. The encoder signal S11 to be supplied to the image processing apparatus 100 is generated.
The microcomputer 140 sets a value corresponding to the height below each thin substrate 704 as the imaging position minimum signal S21 of the imaging position detection means 110, and each thin substrate as the imaging position maximum signal S22 of the imaging position detection means 110. A value corresponding to the height above 704 is set, and the difference between the imaging position minimum signal S21 and the imaging position maximum signal S22 takes into consideration the frame rate and field of view of the imaging means 200 and the moving speed of the lid holding unit 721. Set an appropriate value.

Next, the operation flow of the second embodiment of the present invention will be described with reference to the flowchart shown in FIG.
Step 1: With the thin substrate storage cassette mounted on the fixed base 711, the lifting / lowering operation of the lid holding portion 721 is started.
Step 2: When the lid holding unit 721 reaches a preset reference position and the transmission trigger sensor 410 generates the imaging position initialization signal S12, a position signal S30 generated inside the imaging position detection unit 110 is generated. It is initialized.
Step 3: The identification number N in the processing procedure of the slot 702 held inside the microcomputer 140 is initialized to zero.
Step 4: The microcomputer 140 considers the position of the thin substrate 704 accommodated in the slot 702 corresponding to the identification number N, and sends the imaging position minimum signal S21 and the imaging position maximum signal S22 to the imaging position detection means 110. Instruct.
Step 5: When the lid holding unit 721 moves up and down and the imaging timing of the imaging unit 200 is within the range specified by the imaging position minimum signal S21 and the imaging position maximum signal S22, and the thin substrate 702 is imaged, the imaging position detection unit 110 Updates the imaging position signal S26 and simultaneously outputs an image capture start signal S23.
Step 6: When the image capturing by the image capturing unit 120 is started and the temporary storage of the required amount of images in the image memory 130 is completed, an image capturing completion signal S27 is output to the microcomputer 140.
Step 7: Based on the imaging position signal S26 detected at the imaging timing, the microcomputer 140 performs image processing on the imaging result of the thin substrate 704.
Step 8: The microcomputer 140 adds 1 to the identification number N and updates the value of the identification number N.
Step 9: If the value of the identification number N is the same as the number of slots 702, it is determined that all the slots 702 have been processed, and the processing is completed. If the value of the identification number N is different from the number of slots 702 Is determined to have an unprocessed slot 702, the process returns to step 704, and the same processing is repeated for the next slot 702.

  The present invention provides a workpiece on a conveyor or a thin substrate storage cassette as an object recognition device that can obtain the positional relationship between the imaging means and the workpiece at the imaging timing with high accuracy and obtain a highly accurate image processing result. This can be used for recognizing a thin substrate.

It is the schematic which shows the structure of the 1st Embodiment of this invention. It is a block diagram explaining the 1st Embodiment of this invention. It is a block diagram explaining the imaging position detection means of this invention. It is a timing diagram explaining an embodiment of the present invention. It is the schematic which shows the structure of the 2nd Embodiment of this invention. It is a flowchart explaining the 2nd Embodiment of this invention. It is a figure explaining a prior art. It is a block diagram explaining a prior art.

Explanation of symbols

100: image processing apparatus 110: imaging position detection means 111: vertical synchronization separation circuit 112: counter circuit 113, 114: comparison circuit 115, 116: logic inversion circuit 117: AND circuit 118: latch circuit 120: image capturing means 130: Image memory 140: microcomputer (microcomputer)
200: Imaging means 300: Encoder 310: Linear encoder 400: Trigger sensor 410: Transmission type trigger sensor 500: Conveyor 600: Workpiece 701: Thin substrate storage cassette frame 702: Thin substrate storage cassette slot 703: Thin substrate storage Cassette lid 704: Thin substrate 711: Fixing base 712: Dog 713: Linear scale 721: Cover holder 722: Mounting plate S10: Video signal S11: Encoder signal S12: Imaging position initialization signal S21: Imaging position minimum signal S22: Image pickup position maximum signal S23: Image capture start signal S24, S25: Image signal S26: Image capture position signal S27: Image capture completion signal S30: Position signal S31: Vertical synchronization signal S32: Logic signal output of comparison circuit 113 S33: Comparison circuit 114 The logic of Signal output

Claims (6)

Comprising imaging means for outputting a video signal to image the thin substrate accommodated in a thin substrate storage holder, and an image processing apparatus for processing an imaging result of the previous SL thin substrate, either one of the thin substrate and the imaging means , or the result of both is the imaging of the thin substrate by the imaging means while moving image processing, the object recognition device for recognizing a state of the thin substrate,
A lid opening and closing device for opening and closing the lid of the thin substrate storage cassette;
The fixing part of the lid opening / closing device comprises a fixing base on which the thin substrate storage cassette can be mounted, a dog attached to the fixing base, and a linear scale attached to the fixing base,
The moving unit of the lid opening / closing device holds the lid of the thin substrate storage cassette, and a lid holding unit on which the imaging means and the image processing device are mounted above and below the lid holding unit. An operating mounting plate, a trigger sensor that is mounted on the mounting plate and used to generate an imaging timing of the thin substrate , and a position of the thin substrate and the imaging unit that is mounted on the mounting plate and is imaged by the imaging unit It consists of a linear encoder that detects the relationship ,
The image processing apparatus includes:
The video signal, the imaging position initialization signal, the encoder signal, the imaging position minimum signal, and the imaging position maximum signal, and an image capture start signal and an imaging position detection unit that generates an imaging position signal. ,
When receiving the image capture start signal, the video signal is converted into a digital format and output as an image signal, and when a series of image capture operations are completed, an image capture means for outputting an image capture completion signal;
An image memory for temporarily storing an image signal output from the image capturing means;
A microcomputer for processing the image stored in the image memory using the imaging position signal upon receiving the image capture completion signal;
An object recognition apparatus comprising: The imaging means images the accommodation state of the thin substrate housed in the thin substrate housing cassette as the lid holding unit moves up and down, and supplies the result to the image processing apparatus as a video signal. The object recognition device according to claim 1, The lid holding part holds the lid of the thin substrate storage cassette removed by an opening / closing mechanism by a mechanism such as air adsorption, and moves up and down by a motor driving mechanism or an air driving mechanism. The object recognition apparatus according to 1 . The trigger sensor is configured such that the dog and the detection slit of the trigger sensor intersect at a preset reference position when the lid holding unit moves up and down, and the trigger sensor passes through the dog. at the timing of, and supplying the image pickup position initializing signal to the image processing apparatus, an object recognition system according to claim 1. The linear encoder is configured to intersect with each slit of the linear scale when the lid holding unit is moved up and down, and is in a state every time the linear encoder intersects with each slit of the linear scale. an encoder signal that changes in, and supplying to said image processing apparatus, an object recognition system according to claim 1. The imaging position detection means includes
Using the video signal as an input, a vertical synchronization signal is output by a known vertical synchronization separation circuit,
Using the encoder signal and the imaging position initialization signal as inputs, the relative movement amount of the thin substrate and the imaging means from a preset reference position is generated as a position signal,
It is determined whether the position signal is larger than the imaging position minimum signal and smaller than the imaging position maximum signal, and if the condition is satisfied, the passage of the vertical synchronization signal is permitted,
If the condition is not met, the passage of the vertical synchronization signal is prohibited,
Generate and output an image capture start signal from the resulting signal,
The image capturing start signal is used as a trigger, and the position signal is temporarily stored as an imaging position signal .
The microcomputer sets a value corresponding to the height below each thin substrate as the imaging position minimum signal of the imaging position detection means,
As the imaging position maximum signal of the imaging position detection means, a value corresponding to the height above each thin substrate is set,
The difference between the imaging position minimum signal and the imaging position maximum signal is appropriate in consideration of the frame rate of the imaging means, the field of view of the imaging means, and the relative movement speed of the imaging means and the thin substrate storage cassette. and sets a value, object recognition apparatus according to claim 1.

Images